BRENNEN

Industries / Finance and business services / Finance and banking

Energy trading

Energy trading is the buying and selling of oil, gas and electricity, and of contracts on their future prices, among the producers, users, traders and markets that move them.

1871

In January, 116 dealers, producers, refiners and brokers along Oil Creek in Pennsylvania agreed to form the Titusville Oil Exchange, and on 13 March its members "commenced business" in their new rooms. A Pennsylvania state historical marker calls it "the industry's first permanent oil exchange".

archive.orghmdb.org

39

Milestones achieved across 6 domains.

12

To watch for: announced, not yet confirmed outside.

12

Open deliverables. Nobody has shipped them.

In one minute

Where it started

In 1871 oil men in Titusville, Pennsylvania, formed an exchange, and the oil region's exchanges were soon trading paper certificates for oil held in pipelines. Energy futures began in New York with heating oil in 1978, crude oil in 1983 and natural gas in 1990. Electricity became something to trade when Chile in 1982, and England and Wales in 1990, turned state-run power systems into markets.

Where it stands

Most crude is priced by formula from a few benchmarks that price agencies assess in short daily windows. Power trades on exchanges that are coupled across Europe in 15-minute periods and priced by location in much of the US. The failures, from California in 2000 to oil below zero in 2020 and margin calls in 2022, each showed that a price holds only while the energy can be delivered and the margin can be paid.

What is open

Much of the back office still runs on email, fax and hand checks, and the rules, failures and enforcement cases that shaped the trade sit in scattered documents. The open jobs below build replicas, scorecards and test sets from public or made-up data, so that anyone can check a clearing price, a settlement bill, a forecast or an AI tool. None of them trades real money or gives trading advice.

Domains

AchievedWatch forOpen

Open in energy trading

3 of 12 open deliverables

Open · Physical trading

Planted fakes in cargo papers

A public test set of about 500 made-up document packs for oil cargo deals, each with a sale contract, invoice, bill of lading and inspection certificate. Some packs carry planted frauds of the kinds seen in real cases: a sale the buyer never agreed to, one cargo financed twice with two lenders, a bill of lading copied from another pack, and quantities that do not add up. Each pack has an answer sheet, and a scoring script reports how many fakes a checking tool catches and how many honest packs it flags by mistake. Every party, ship and amount is invented.

Built with what exists

Python to generate packs from a list of fields; a PDF library and scan-noise filters for realistic copies; AI models such as Claude, GPT and Gemini, and open models such as Qwen, as the first tools to score.

Credit · unclaimed

Open · Power markets

Replicate the Iberian auction

An extension of OpenEUPHEMIA, an open-source copy of Europe's day-ahead clearing algorithm, to Spain and Portugal. It rebuilds one month of the Iberian exchange OMIE's day-ahead auction from OMIE's public bid files and scores the result against OMIE's published prices.

Built with what exists

Python; OpenEUPHEMIA (MIT licence); the HiGHS open-source solver; OMIE's public files; an AI coding agent such as Claude Code to parse OMIE's formats and track down hours that do not match.

Credit · unclaimed

Open · Forecasting

AI weather scored in megawatts

A daily public scorecard for three to five European bidding zones that turns open weather forecasts, from ECMWF's AIFS and its physics model IFS to other open AI models, into day-ahead wind and solar output forecasts in MW. It scores each against actual output, side by side with the official forecast that each grid operator must publish by 18:00 Brussels time the day before.

Built with what exists

AIFS open data from ECMWF; the ENTSO-E Transparency Platform for the grid operators' forecasts and actual output; Python with gradient-boosted models to turn weather into MW; one rented GPU to run open AI weather models.

Credit · unclaimed

Where it sits